Diagnostic significance of prostate-specific antigen velocity at intermediate PSA serum levels in relation to the standard deviation of different test systems.

Serial prostate-specific antigen (PSA) measurements (PSA velocity) as an additional instrument to detect prostatic cancer was introduced in 1992. It has previously been reported that PSA increase per year differed in the last 5 years prior to diagnosis in patients with benign prostatic hyperplasia (0.18 ng/ml/year), locally confined (0.75 ng/ml/year) and metastasized (4.4 ng/ml/year) cancer of the prostate (CaP) in contrast to healthy men (0.04 ng/ml/year). The ability of PSA velocity to detect organ-confined CaP in patients with intermediate PSA serum values depends therefore on a reliable and reproducible PSA result. The present study comprised 85 men with PSA values between 3 and 8 ng/ml (Abbott IMx). PSA measurements were repeated with Abbott IMx (n = 85 patients) and Hybritech Tandem-E (n = 59 patients) assays. The PSA serum values differed from one examination to the other from 0.02 to 2.74 ng/ml with the Abbott IMx. Standard deviation amounted to 0.35 ng/ml with the Abbott IMx PSA assay. Using the Hybritech Tandem-E assay, mean standard deviation was 1.15 ng/ml and therefore higher than with the Abbott IMx assay. The difference from one test to the other ranged from 0.05 to 4.05 ng/ml with the Hybritech Tandem-E. Using the Abbott IMx assay, 10.6% of all repeat measurements exceeded 1 ng/ml whereas in the Hybritech Tandem-E assay 62.7% of the second measurements differed > 1 ng/ml from the first PSA result. An increase in PSA serum values may therefore be due to intratest variation, physiological day-to-day variation as well as prostatic disease. It is important to notice that the intra-assay variation may be greater than the PSA increase per year in a patient with CaP. Therefore, PSA velocity seems to be of limited value.